Intelligent shelf and AGV cooperative operation path planning method

Abstract

The present application relates to the field of warehouse management, and more particularly to a path planning method for intelligent shelves and AGV collaborative operation. It comprises: based on the real-time data of intelligent shelves and AGV in the warehouse management system, the priority attraction of each intelligent shelf is dynamically calculated for each AGV through the shelf priority attraction calculation algorithm; according to the priority attraction of the intelligent shelf calculated in real time, the intelligent shelf with the highest priority attraction is selected as the target intelligent shelf; based on the target intelligent shelf, the shortest path is planned, and the path conflict is avoided through time window scheduling and priority rules. The technical problem that the AGV usually only selects the target intelligent shelf based on the distance or fixed rules, and fails to comprehensively consider the inventory status of the intelligent shelf, the matching of the commodity category and the task type, and the dynamic mutual influence of the inventory between the intelligent shelves of the same commodity category, which may lead to uneven task allocation or resource waste, is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of warehouse management, and in particular to a path planning method for intelligent shelves and AGV collaborative operation. BACKGROUND

[0002] With the rapid development of the logistics industry, intelligent warehouse technology has gradually become a key technology to improve warehouse efficiency and reduce operating costs. Traditional warehouse systems mostly rely on manual operation and fixed path planning, which not only easily leads to low efficiency and human error, but also cannot respond to dynamic changes in complex environments in real time. With the continuous development of intelligent logistics and automated warehouse systems, the path planning of intelligent shelves and automated guided vehicles (AGV) collaborative operation has become a key problem in modern warehouse management systems. Intelligent shelf systems can improve inventory management efficiency and warehouse space utilization by accurately storing and managing goods, while AGV, as an automated transportation tool, can realize autonomous material handling, reduce manual operation, and improve work efficiency and safety.

[0003] With the continuous development of intelligent shelf and AGV collaborative operation technology, the research on path planning method has become a key to improve warehouse efficiency and reduce operating costs. By combining intelligent shelf systems with AGV path planning, efficient task scheduling and path optimization can be achieved in dynamic environments, providing strong technical support for the development of automated warehouse systems.

[0004] However, the above-mentioned existing path planning method for intelligent shelves and AGV collaborative operation still has the following technical problems: Traditional warehouse systems often base on static rules or simple distance priority in AGV task allocation and path planning, which is difficult to adapt to complex scenarios of multi-category goods and task surge in peak periods, leading to AGV empty, detour or invalid movement, and low overall work efficiency; Lack of real-time response to real-time changes in inventory status, task demand and AGV position in the warehouse environment, which easily leads to unreasonable task allocation or path conflict, affecting system throughput; When selecting intelligent shelves, only distance factor is considered, ignoring the inventory status of intelligent shelves, commodity category matching and the collaborative effect of neighboring intelligent shelves, which leads to the selection of suboptimal target intelligent shelves by AGV, increasing the moving distance and time cost. SUMMARY

[0005] The present application provides a path planning method for intelligent shelves and AGV collaborative operation to solve the technical problems that AGV usually selects target intelligent shelves based on distance or fixed rules only, without considering the inventory status of intelligent shelves, commodity category matching and task type; It is difficult to use the remaining capacity or inventory data of intelligent shelves in real time, which may lead to the selection of full or empty intelligent shelves by AGV; The dynamic mutual influence between intelligent shelves of the same commodity category is not considered, which may lead to uneven task allocation or resource waste.

[0006] The application discloses a path planning method for intelligent shelves and AGVs to cooperate, and specifically comprises the following technical solutions.

[0007] The application discloses a path planning method for intelligent shelves and AGVs to cooperate, and specifically comprises the following technical solutions.

[0008] S1. Based on real-time data of intelligent shelves and AGVs in a warehouse management system, a shelf priority attraction calculation algorithm is used to dynamically calculate the priority attraction of each intelligent shelf for each AGV.

[0009] S2. According to the priority attraction of the intelligent shelf calculated in real time, the intelligent shelf with the highest priority attraction is selected as a target intelligent shelf; based on the target intelligent shelf, a shortest path is planned, and path conflicts are avoided through time window scheduling and priority rules.

[0010] Preferably, the S1 specifically comprises the following.

[0011] In the implementation process of the shelf priority attraction calculation algorithm, the task priority of each AGV and each intelligent shelf is calculated, and the task priority reflects whether the intelligent shelf is suitable for the current task of the AGV based on the commodity category matching and the inventory state constraint.

[0012] Preferably, the S1 specifically comprises the following.

[0013] In the implementation process of the shelf priority attraction calculation algorithm, when the intelligent shelf meets the commodity category matching and the inventory state constraint, the task priority is set as a basic task priority value, otherwise, the task priority is set as zero, indicating that the intelligent shelf has no attraction to the AGV.

[0014] Preferably, the S1 specifically comprises the following.

[0015] In the implementation process of the shelf priority attraction calculation algorithm, a dynamic capacity is introduced and defined according to the task type of the AGV: for a delivery task, the dynamic capacity is equal to the remaining capacity of the intelligent shelf, reflecting the ability of the intelligent shelf to receive commodities; for a pick-up task, the dynamic capacity is equal to the inventory of the intelligent shelf, reflecting the ability of the intelligent shelf to provide commodities.

[0016] Preferably, the S1 specifically comprises the following.

[0017] In the implementation process of the shelf priority attraction calculation algorithm, the dynamic capacity is divided by the maximum storage capacity of the intelligent shelf to obtain a normalized proportion value, the normalized proportion value is added by 1 to form an amplification factor; and a distance attenuation term is introduced to reflect that the closer the distance between the intelligent shelf and the AGV, the smaller the priority attraction of the intelligent shelf, and the path length is optimized.

[0018] Preferably, the S1, specifically includes:

[0019] In the implementation process of the shelf priority attraction calculation algorithm, the cooperative effect of adjacent intelligent shelves is introduced to reflect the mutual influence of the inventory dynamics between intelligent shelves; the direction of the cooperative effect is determined by combining an exponential decay function and the product of the normalized neighbor intelligent shelf inventory change and the AGV task type based on a sign function, and the priority attraction of the intelligent shelf is calculated in real time to dynamically optimize the selection of the intelligent shelf by the AGV.

[0020] Preferably, the S2, specifically includes:

[0021] The time window scheduling assigns a time window to the path of the AGV, records the time occupation interval of each path point, and when two paths occupy the same path point in the same time window, the path of the low-priority AGV is adjusted and the path of the high-priority AGV is selected.

[0022] Preferably, the S2, specifically includes:

[0023] The priority rule is based on the basic task priority value of the AGV to determine the priority in the path conflict, and the high-priority AGV retains the original path and the low-priority AGV re-plans the path.

[0024] Preferably, the S2, specifically includes:

[0025] During the execution of the AGV task, when the inventory change and the path conflict problem are checked, the priority attraction calculation of the intelligent shelf is re-executed, and the target intelligent shelf is updated, and the A* algorithm is used to re-plan the path according to the updated target intelligent shelf.

[0026] The beneficial effects of the technical scheme of the application are:

[0027] 1. By introducing the shelf priority attraction calculation algorithm, the real-time inventory state of the intelligent shelf, the commodity category matching, the task type, the distance factor and the cooperative effect of the neighbor intelligent shelf are comprehensively considered to dynamically generate the priority attraction of each intelligent shelf for each AGV, which breaks through the limitation of traditional static task allocation, can accurately match the task demand of the AGV according to the real-time change of the warehouse environment (such as inventory quantity, remaining capacity), significantly shortens the task response time of the AGV by selecting the intelligent shelf with the highest priority attraction as the target intelligent shelf, reduces the empty or invalid movement, thereby improving the overall completion efficiency of the warehouse task, and is especially suitable for multi-category goods and complex scenes during peak period.

[0028] 2、Through multi-dimensional evaluation such as commodity category matching, inventory state constraint, dynamic capacity normalization, distance attenuation and synergistic effect, the rationality of intelligent shelf selection is ensured, the utilization rate of intelligent shelf resources is optimized, invalid task allocation caused by insufficient inventory or capacity saturation is avoided, resource waste of the warehouse system is reduced, and the economy and sustainability of the operation are improved.

[0029] 3、The synergistic effect of neighbor intelligent shelves is introduced, the priority attraction of the intelligent shelf is dynamically adjusted by analyzing the inventory change trend and distance relationship of the neighbor intelligent shelves, the dynamic correlation between the intelligent shelves can be perceived, the scene of rapid inventory change can be adapted, such as order surge in peak period, so as to improve the flexibility and accuracy of task allocation, and reduce task interruption or re-planning caused by local inventory fluctuation.

[0030] 4、By assigning basic task priority values to different tasks, such as higher priority for urgent orders than for ordinary orders, and by prioritizing the path of high-priority AGVs when there is a path conflict, the system can quickly respond to urgent task demands, optimize the processing capacity of the warehouse management system for high-value or time-sensitive orders, improve customer satisfaction and operational efficiency, and is particularly suitable for e-commerce, cold chain logistics and other scenarios with high time efficiency requirements. BRIEF DESCRIPTION OF DRAWINGS

[0031] Figure 1 A flowchart of the path planning method for the intelligent shelf and AGV collaborative operation according to the present application. DETAILED DESCRIPTION

[0032] In order to further illustrate the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0033] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0034] The specific scheme of the path planning method for the intelligent shelf and AGV collaborative operation provided by the present application will be described in detail below in conjunction with the drawings.

[0035] Referring to the drawings Figure 1 , which shows a flowchart of the path planning method for the intelligent shelf and AGV collaborative operation provided by an embodiment of the present application. The method comprises the following steps:

[0036] S1, based on the real-time data of intelligent shelves and AGVs in the warehouse management system, the priority attraction of each intelligent shelf is dynamically calculated for each AGV through a shelf priority attraction calculation algorithm;

[0037] The warehouse environment is a two-dimensional layout, including intelligent shelves, AGVs, channels, and possible static obstacles (such as walls or equipment), etc., represented by a grid map for AGV path planning;

[0038] There are fixed-position intelligent shelves in the warehouse management system, and the intelligent shelf position set is represented as , where the position of each intelligent shelf is determined by two-dimensional coordinates (unit: meters), and the intelligent shelf is equipped with sensors and communication modules to monitor and report inventory status in real time, such as inventory quantity and remaining capacity;

[0039] There are AGVs deployed in the warehouse management system, and the AGV position set is , where the position of each AGV is represented by two-dimensional coordinates (unit: meters), and the AGV is responsible for executing delivery (delivering goods to designated intelligent shelves) or picking (picking goods from intelligent shelves) tasks, equipped with a positioning system, a path planning module, and communication equipment;

[0040] Based on the real-time data of intelligent shelves and AGVs in the warehouse management system, including intelligent shelf position, inventory quantity, remaining capacity, AGV position, task type (delivery or picking), task priority, and commodity category, the priority attraction of each intelligent shelf in the warehouse management system is dynamically calculated for each AGV through a shelf priority attraction calculation algorithm, serving as the basis for path planning;

[0041] The shelf priority attraction calculation algorithm calculates the task priority for each AGV and each intelligent shelf, and the task priority reflects whether the intelligent shelf is suitable for the current task of the AGV based on two conditions, which are: commodity category matching and inventory state constraint; for commodity category matching, by checking whether the predefined commodity category of the intelligent shelf is consistent with the commodity category of the current task of the AGV, only when the two match, the intelligent shelf can be attractive to the AGV, otherwise, it is not attractive; for inventory state constraint, it needs to be processed according to the task type of the AGV: if the AGV performs a delivery task, it needs to check whether the remaining capacity of the intelligent shelf is greater than zero, that is, the intelligent shelf is not full, and the intelligent shelf with zero remaining capacity cannot receive goods, that is, the AGV is not suitable for performing a delivery task, otherwise, the AGV is suitable for performing a delivery task; if the AGV performs a pick-up task, it checks whether the inventory of the intelligent shelf is greater than zero, that is, the intelligent shelf is not empty, and the intelligent shelf with zero inventory cannot provide goods, that is, the AGV is not suitable for performing a pick-up task, otherwise, the AGV is suitable for performing a pick-up task; if the intelligent shelf satisfies the commodity category matching and the inventory state constraint, the task priority is assigned to the basic task priority value, otherwise, the task priority is set to zero, indicating that the intelligent shelf has no attraction to the AGV, which is expressed in the formula as follows:

[0042]

[0043] represents the intelligent shelf to the task priority of the AGV , introduces the task priority, such as the emergency order priority being higher than the ordinary order, to improve the response ability to high-priority tasks; represents the basic task priority value of the AGV , which is assigned based on the urgency or importance of the task, such as an emergency order , an ordinary order , which can be set according to specific implementation scenarios, and is not limited here; represents the predefined commodity category of the intelligent shelf , including food, electronic products, clothing, etc.; represents the commodity category of the current task of the AGV ; represents the dynamic capacity;

[0044] The shelf priority attraction calculation algorithm introduces dynamic capacity to quantify the suitability of the intelligent shelf to the AGV task, which is defined according to the task type of the AGV: for a delivery task, the dynamic capacity is equal to the remaining capacity of the intelligent shelf, reflecting the ability of the intelligent shelf to receive goods; for a pick-up task, the dynamic capacity is equal to the inventory of the intelligent shelf, reflecting the ability of the intelligent shelf to provide goods, which is defined as follows:

[0045]

[0046] wherein, represents the dynamic capacity, which is derived from the real-time inventory status of the smart shelf, according to the task type of the AGV , the remaining capacity or the inventory of the smart shelf is selected to ensure that the priority attraction of the smart shelf matches the task demand of the AGV; represents the remaining capacity of the smart shelf at time ; represents the inventory of the smart shelf at time ; represents the task type of the AGV , represents that the AGV performs a delivery task, represents that the AGV performs a pick-up task;

[0047] The dynamic capacity is further divided by the maximum storage capacity of the smart shelf to obtain a normalized proportion value, which is in the range of 0 to 1. In order to enhance the sensitivity of the priority attraction of the smart shelf to the capacity status of the smart shelf, the normalized proportion value is added by 1 to form an amplification factor, which is in the range of 1 to 2, to ensure that the smart shelf with larger remaining capacity or inventory has higher attraction to the corresponding task of the AGV;

[0048] In order to reflect the influence of the distance between the smart shelf and the AGV on the priority attraction of the smart shelf, a distance attenuation term is introduced, which is calculated by dividing the Euclidean distance between the current position of the AGV and the position of the smart shelf by the maximum distance between the AGV and all positions of the smart shelf, and then multiplied by a distance weight parameter, and added to prevent division by zero, to reflect that the farther the distance between the smart shelf and the AGV, the larger the denominator, the smaller the priority attraction of the smart shelf, to ensure that the AGV tends to select a closer smart shelf, thereby optimizing the path length;

[0049] In order to further optimize the selection of the smart shelf by the AGV, the shelf priority attraction calculation algorithm introduces the synergistic effect of adjacent smart shelves to reflect the mutual influence of the inventory dynamics between smart shelves, determines a set of neighbor smart shelves, i.e. smart shelves with the same predefined product category as the current smart shelf, for each neighbor smart shelf, the distance between the current smart shelf and the neighbor smart shelf is calculated, and an exponential decay function is used to reflect that the closer the distance between the current smart shelf and the neighbor smart shelf, the larger the synergistic effect weight, and the farther the distance between the current smart shelf and the neighbor smart shelf, the synergistic effect weight tends to zero;

[0050] For delivery task, i.e. task type is positive, if the neighbor smart shelf inventory increases, i.e. , it indicates that the current smart shelf storage demand rises, and the synergy effect is positive, represents the inventory change of the neighbor smart shelf, i.e. the current inventory of the neighbor smart shelf minus the inventory at the previous time; if the neighbor smart shelf inventory decreases, i.e. , it indicates that the current smart shelf storage demand falls, and the synergy effect is negative; if the inventory change of the neighbor smart shelf is zero, the synergy effect is zero; for taking task, i.e. task type is negative, if the neighbor smart shelf inventory decreases, i.e. , it indicates that the current smart shelf taking demand rises, and the synergy effect is positive; if the neighbor smart shelf inventory increases, i.e. , it indicates that the current smart shelf taking demand falls, and the synergy effect is negative; if the inventory change of the neighbor smart shelf is zero, the synergy effect is zero;

[0051] The calculation formula of the priority attraction of the smart shelf to the AGV is

[0052]

[0053] wherein, represents the priority attraction of the smart shelf to the AGV at time , reflecting the priority of the smart shelf to the AGV to execute the delivery or taking task, the greater the value of the priority attraction, the more suitable the smart shelf is selected as the target smart shelf by the AGV ; represents the basic attraction of the smart shelf to the AGV ; represents the normalized dynamic capacity, used to measure the proportion of the remaining capacity or inventory of the smart shelf relative to the maximum capacity of the smart shelf, which is when executing the delivery task, the smart shelf with the greater remaining capacity is preferentially selected, the higher the proportion of the remaining capacity of the smart shelf relative to the maximum capacity of the smart shelf, the greater the priority attraction, which is when executing the taking task, the smart shelf with the greater inventory is preferentially selected, the higher the proportion of the inventory of the smart shelf relative to the maximum capacity of the smart shelf, the greater the priority attraction; represents the maximum storage capacity of the smart shelf , unit: pieces, i.e. the maximum capacity of the smart shelf, reflecting the total amount of goods that the smart shelf can accommodate; represents the distance decay term, reflecting the distance between the smart shelf and the AGV The influence of the distance between the AGV and the smart shelf on the smart shelf's priority attraction decreases as the distance increases, and the AGV prefers to choose the smart shelf that is closer to it, The influence of the distance between the AGV and the smart shelf on the smart shelf's priority attraction decreases as the distance increases, and the AGV prefers to choose the smart shelf that is closer to it, The influence of the distance between the AGV and the smart shelf on the smart shelf's priority attraction decreases as the distance increases, and the AGV prefers to choose the smart shelf that is closer to it, The influence of the distance between the AGV and the smart shelf on the smart shelf's priority attraction decreases as the distance increases, and the AGV prefers to choose the smart shelf that is closer to it, To ensure that the denominator is not zero, avoid the problem of division by zero; represents the maximum distance between the AGV and all smart shelf positions, used for normalization processing; represents the distance weight parameter, used to adjust the degree of attenuation of the distance on the priority attraction of the smart shelf, which can be set according to the specific implementation scene, and is not limited here, and is recommended ; represents the synergy effect part, which is based on the inventory changes of the neighboring smart shelves and the task type of the AGV, and dynamically adjusts the priority attraction of the smart shelf ; represents the synergy effect part, which is based on the inventory changes of the neighboring smart shelves and the task type of the AGV, and dynamically adjusts the priority attraction of the smart shelf ; represents the sum of all smart shelves in the neighbor shelf set of the smart shelf (not including the shelf itself), accumulates the synergy effect contribution of all neighboring smart shelves, and comprehensively reflects the influence of the state of the neighboring smart shelves on the priority attraction of the smart shelf ; represents the sum of all smart shelves in the neighbor shelf set of the smart shelf (not including the shelf itself), accumulates the synergy effect contribution of all neighboring smart shelves, and comprehensively reflects the influence of the state of the neighboring smart shelves on the priority attraction of the smart shelf ; represents the neighbor shelf set of the smart shelf, i.e. the smart shelves with the same predefined commodity category as the smart shelf ; represents the synergy effect weight, used to adjust the influence degree of the neighboring smart shelves on the priority attraction of the smart shelf, and controls the strength of the synergy effect, which can be set according to the specific implementation scene, and is not limited here, and is recommended ; represents the synergy effect weight, used to adjust the influence degree of the neighboring smart shelves on the priority attraction of the smart shelf, and controls the strength of the synergy effect, which can be set according to the specific implementation scene, and is not limited here, and is recommended ; represents the exponential decay function, based on the distance between the smart shelf and the neighboring smart shelf , so that the synergy effect decays exponentially with the distance , the synergy effect of the smart shelf in the near distance is stronger, represents the maximum distance between the smart shelf and the neighboring smart shelf , used for normalization processing; represents the maximum distance between the smart shelf and the neighboring smart shelf , used for normalization processing; represents the maximum distance between the smart shelf and the neighboring smart shelf , used for normalization processing; represents the maximum distance between the smart shelf and the neighboring smart shelf , used for normalization processing; represents the maximum distance between the smart shelf and the neighboring smart shelf , used for normalization processing;represents a picking task; represents a neighbor smart shelf represents the inventory change amount of the neighbor smart shelf, i.e. the current inventory amount of the neighbor smart shelf minus the inventory amount at the previous time; represents the maximum capacity of the neighbor smart shelf represents the maximum capacity of the neighbor smart shelf

[0054] By calculating the priority attraction of the smart shelf in real time, the selection of the smart shelf by the AGV is dynamically optimized, the path length of the AGV is shortened, the empty or invalid movement is reduced, the task completion efficiency is improved, and it is especially suitable for complex warehouse scenes with multiple categories and peak periods;

[0055] S2, according to the priority attraction of the smart shelf calculated in real time, selecting the smart shelf with the highest priority attraction as the target smart shelf; based on the target smart shelf, planning the shortest path, and avoiding path conflicts through time window scheduling and priority rules;

[0056] According to the priority attraction of the smart shelf calculated in real time, the smart shelf with the highest priority attraction is selected as the target smart shelf, which is expressed by the formula:

[0057]

[0058] Among them, represents the number of the target smart shelf of the AGV at the time ; represents selecting the smart shelf with the maximum priority attraction from smart shelves ; represents the total number of smart shelves;

[0059] Using A* algorithm to plan the shortest path according to the current position of the AGV to the position of the target smart shelf, while avoiding path conflicts through time window scheduling and priority rules; the A* algorithm is a well-known technical means to those skilled in the art, and will not be described here; the time window scheduling assigns a time window to the path of the AGV , records the time occupation interval of each path point, if two paths occupy the same path point in the same time window, i.e. potential collision occurs, adjust the path of the AGV with low priority or delay the departure time, select the path of the AGV with high priority; the priority rule determines the priority in path conflict based on the basic task priority value of the AGV , the AGV with high priority keeps the original path, and the AGV with low priority re-plans the path;

[0060] If the inventory changes are detected during path execution or the path needs to be re-planned due to path conflicts, the current path of the AGV is interrupted, the priority attraction calculation of the intelligent shelf is re-executed, the target intelligent shelf is updated, and then the path is re-planned according to the updated target intelligent shelf using the A* algorithm, so as to ensure that the AGV always moves towards the current optimal intelligent shelf and the path is safe;

[0061] Through dynamic path adjustment, the AGV path selection is optimized, the invalid movement is reduced, and the operation efficiency is improved.

[0062] In summary, a path planning method for collaborative operation of intelligent shelves and AGVs is completed.

[0063] The order of the embodiments of the application is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are possible or can be advantageous.

[0064] Each embodiment in the specification is described in a progressive manner, and the same or similar parts between each embodiment can be referred to each other. Each embodiment mainly describes the difference from other embodiments.

[0065] The above embodiments are only used to illustrate the technical solutions of the present application, but not limit the same. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that the technical solutions recorded in the foregoing embodiments can be modified, or some technical features can be replaced by equivalent ones. The modification or replacement does not change the essence of the corresponding technical solution, and should be included in the protection scope of the present application.

Claims

1.A method for path planning of intelligent shelves and AGVs, characterized in that, Comprising the following steps: S1. Based on the real-time data of intelligent shelves and AGVs in the warehouse management system, the priority attraction of each intelligent shelf for each AGV is dynamically calculated by a shelf priority attraction calculation algorithm; in the implementation process of the shelf priority attraction calculation algorithm, the dynamic capacity is introduced to calculate the task priority of the intelligent shelf for the AGV, and the task priority reflects whether the intelligent shelf is suitable for the current task of the AGV based on the commodity category matching and the inventory state constraint; the dynamic capacity is divided by the maximum storage capacity of the intelligent shelf to obtain a normalized proportion value, and the normalized proportion value is added by 1 to form an amplification factor; and a distance attenuation term is introduced, the Euclidean distance between the current position of the AGV and the position of the intelligent shelf is calculated, divided by the maximum distance between the AGV and all intelligent shelf positions, multiplied by the distance weight parameter after division, and added by 1 to prevent division by zero, reflecting that the farther the distance between the intelligent shelf and the AGV, the smaller the priority attraction of the intelligent shelf, and the optimization path length; the priority attraction of the intelligent shelf is dynamically adjusted by introducing the synergy effect of adjacent intelligent shelves; The formula for the priority attraction of the intelligent shelf to the AGV is wherein, denotes an intelligent shelf at time the priority attraction of the AGV ; denotes an intelligent shelf the task priority of the AGV ; denotes the dynamic capacity; denotes the maximum storage capacity of the intelligent shelf ; denotes the distance between the intelligent shelf and the AGV ; denotes the maximum distance of the AGV to all intelligent shelf locations; denotes the distance weight parameter; denotes the neighbor shelf set of the intelligent shelf ; denotes the synergy effect weight; denotes the distance of the intelligent shelf to the neighbor intelligent shelf ; denotes the maximum distance of the intelligent shelf to the neighbor intelligent shelf ; denotes the sign function; denotes the inventory change amount of the neighbor intelligent shelf ; denotes the AGV task type; denotes the maximum storage capacity of the intelligent shelf ; S2. According to the real-time calculated priority attraction of the intelligent shelf, the intelligent shelf with the highest priority attraction is selected as the target intelligent shelf; based on the target intelligent shelf, the shortest path is planned, and path conflicts are avoided through time window scheduling and priority rules. 2.The path planning method for the cooperation between the intelligent shelf and the AGV according to claim 1, characterized in that, The S1 specifically comprises: In the implementation process of the shelf priority attraction calculation algorithm, when the intelligent shelf meets the commodity category matching and the inventory state constraint, the task priority is given as the basic task priority value, otherwise the task priority is set to zero, indicating that the intelligent shelf has no attraction to the AGV; for commodity category matching, check whether the pre-defined commodity category of the intelligent shelf is consistent with the commodity category of the current task of the AGV, only when the two match, the intelligent shelf has attraction to the AGV, otherwise it has no attraction; for inventory state constraint, when the AGV performs a delivery task, it needs to check whether the remaining capacity of the intelligent shelf is greater than zero, if the remaining capacity is zero, the AGV is not suitable for performing the delivery task, otherwise the AGV is suitable for performing the delivery task; when the AGV performs a pick-up task, check whether the inventory of the intelligent shelf is greater than zero, if the inventory is zero, the AGV is not suitable for performing the pick-up task, otherwise the AGV is suitable for performing the pick-up task. 3.The path planning method for the cooperation between the intelligent shelf and the AGV according to claim 2, characterized in that, The S1 specifically comprises: In the implementation process of the shelf priority attraction calculation algorithm, the dynamic capacity is introduced, which is defined according to the task type of the AGV: for a delivery task, the dynamic capacity is equal to the remaining capacity of the intelligent shelf, reflecting the ability of the intelligent shelf to receive goods; for a pick-up task, the dynamic capacity is equal to the inventory of the intelligent shelf, reflecting the ability of the intelligent shelf to provide goods. 4.The path planning method for intelligent shelves and AGVs to work cooperatively according to claim 1, characterized in that, The S1 specifically comprises: In the implementation process of the shelf priority attraction calculation algorithm, the cooperative effect of adjacent intelligent shelves is introduced to reflect the mutual influence of inventory dynamics between intelligent shelves, and the set of neighbor intelligent shelves is determined; for each neighbor intelligent shelf, the distance between the current intelligent shelf and the neighbor intelligent shelf is calculated, and the direction of the cooperative effect is determined by combining the exponential decay function with the product of the normalized neighbor intelligent shelf inventory change and the AGV task type based on the sign function, the priority attraction of the intelligent shelf is calculated in real time, and the selection of the intelligent shelf by the AGV is dynamically optimized. 5.The path planning method for intelligent shelves and AGVs to work cooperatively according to claim 1, characterized in that, The S2 specifically comprises: The time window scheduling assigns a time window to the path of the AGV, records the time occupation interval of each path point, adjusts the path of the low-priority AGV when two paths occupy the same path point in the same time window, and selects the path of the high-priority AGV. 6.The path planning method for the cooperation between the intelligent shelf and the AGV according to claim 5, characterized in that, The S2 specifically comprises: The priority rule is based on the priority value of the basic task of the AGV to determine the priority in the path conflict, the high-priority AGV retains the original path, and the low-priority AGV replans the path. 7.The path planning method for the cooperation between the intelligent shelf and the AGV according to claim 6, characterized in that, The S2 specifically comprises: During the execution of the AGV task, when the inventory change and the path conflict problem are checked, the priority attraction calculation of the intelligent shelf is re-executed, and the target intelligent shelf is updated, and the path is replanned using the A* algorithm according to the updated target intelligent shelf.

Images